Phishing involves manipulating individuals into revealing private data, e.g., user IDs, bank details, and passwords. The observed surge in fraud is related to increased deception, impersonation, and advanced online attacks. Thus, effective phishing detection methods are required to mitigate escalating global phishing threats. Existing methods (e.g., heuristics-based, signature-based, and visual similarity-based methods) attempt to detect phishing sites, and machine learning (ML) and deep learning (DL) methods are effective in the cybersecurity context in terms of learning from data, offering insights, and forecasting. However, independent ML algorithms are limited when handling complex data, and DL techniques surpass traditional ML methods in terms of performance but require more data and time. To tackle these challenges, we present EnLeM, an ensemble learning model designed specifically for phishing website detection. EnLeM brings together three well-known machine learning classifiers—decision tree, random forest, and k-nearest neighbor—using a hard voting mechanism, and further strengthens efficiency with Mutual Information–based feature selection. When tested on the UCI phishing dataset, EnLeM delivered strong results, reaching 97.21% accuracy and a 97.51% F1-score. Compared to individual ML classifiers, it consistently performed better, and it also proved more efficient than deep learning models such as CNN and LSTM. Notably, EnLeM maintained stable accuracy across different feature subsets while cutting execution time by roughly 13%. By striking a balance between accuracy, speed, and interpretability, EnLeM stands out as a practical and scalable solution for real-time phishing detection without the heavy resource demands of deep learning approaches.

Electrical Impedance Tomography based sensors have emerged as a promising approach in tactile sensing, offering notable advantages such as affordability, portability, and low power consumption. However, the inherently ill-posed nature of the inverse problem often results in reconstruction errors, impacting on the accuracy of tactile information retrieval. In this work, an effective deep learning approach for tactile sensing is proposed, leveraging the concept of learnable masks, incorporated within a Convolutional Neural Network. The learnable masks support the selection of the most informative feature subsets from the associated voltage inputs, enabling the network to reconstruct conductivity distributions precisely. The proposed approach exhibited outstanding performance in image reconstruction, achieving a mean square error of 0.000041, a structural similarity index of 98.28, and a peak signal-to-noise ratio of 42.35 dB.

Hardware Trojans (HTs) have emerged as a security threat to the integrated circuits (ICs) industry. To counteract this threat, various detection methods have been proposed, among which conventional machine learning (ML)-based techniques using heuristic features from gate-level netlists have gained wide acceptance. However, these methods are notably sensitive to minor perturbations in modifications of the test circuits, often resulting in decreased detection capabilities. Furthermore, the black-box nature of the ML models obscures the basis for their decisions. This lack of transparency makes it difficult to scrutinize and address potential flaws in the models, thereby further reducing the credibility of Trojan detection results. In response to these challenges, we propose a targeted solution that leverages the SHapley Additive exPlanations method (SHAP), which dismantles the black-box paradigm and clarifies the fundamental reasons behind the failure of existing detection methods under circuit sample perturbations. Building on these insights, we abandon classical ML-based detection in favor of a scheme based on graph learning (GL), which significantly reduces the average drop in Recall from 52.35% to 7.29% compared with the traditional method. Comparative experiments demonstrate that our proposed GL-based method effectively resolves the sensitivity issue related to the sample perturbations in existing HT detection approaches.

Rapid technological advances have led to profound changes to skills composition in the workplace. Low skilled jobs are gradually being replaced by automated systems, while there is a gaining demand for jobs which require interpersonal and technological skills. In this study, a combination of data science techniques are used to study this phenomenon. Firstly, matrix factorization and clustering methods are used to extract skills dimensions from O-NET, a database of occupations-skills matchings compiled by the US Department of Labor. A method of evaluating the performance of each of these methods is proposed and used to determine the ideal extraction procedure. Next, the relative importance of different occupations was estimated using a corpus of job advertisements collected from local job sites, which we adopt as a proxy for demand. Finally, the results of this analysis are used to measure shifts in the demand for the skills and abilities associated with these occupations. This procedure is applicable to any job market. However as a test of its effectiveness, it was used to study two important economic sectors in the United Arab Emirates (UAE): Oil & Gas and Banking & Finance. The findings of this study will help us to determine the jobs and skills which will be most impacted by structural and technological change, and provide recommendations to ensure that the UAE workforce is well equipped to adapt to these changes.

We demonstrate an integrated strategy for identifying buildings in very high resolution satellite imagery of urban areas. Buildings are extracted using structural, contextual, and spectral information. We perform multi-resolution and spectral difference segmentation to obtain a proper object segmentation. First, we use One-Class support vector machine (SVM) in order to determine the man-made structures (buildings, roads, etc.). Next, we proceed with texture segmentation approach using a conditional threshold value to extract the buildings. And then, we use geodesic opening and closing operations to extract bright foreground objects. After this, shadows and vegetation regions are detected in these segments based on their spectral properties. We then remove noise, vegetation and shadows from the candidate building regions. And finally, we classify the buildings by checking for the presence of shadows along the buildings opposite to the sun's azimuth direction to distinguish buildings from other bright regions. Performance evaluation of the proposed algorithm is performed on data acquired using WorldView satellite imagery over Abu Dhabi, United Arab Emirates.

Abstract Accuracy is still the greatest challenge in the wireless sensor network localization efforts. Several diverse factors can give rise to localization errors. Modeling such diverse influencing factors to deliver a single, reasonably simple and practical solution is a difficult task. In order to address the problem of location inaccuracy, we propose a comparatively simple and ingenious approach, which is the simultaneous perturbation stochastic approximation (SPSA) localization engine. SPSA bypasses tedious modeling of the influencing factors where some of them are yet to be explored and random in nature. SPSA‐based localization estimates the non‐anchor node locations through minimizing the summation of estimated errors of all neighbors. However, the downside of SPSA is that it incurs errors in some specific relative neighborhood configurations often referred to as flip ambiguity. So, we further propose a solution to the flip ambiguity problem by implementing a constrained optimization with a penalty function method on the identified flip nodes. Most importantly, error propagation of the iterative localization algorithm is managed by incorporating a neighbor confidence matrix. We name this modified SPSA engine as simultaneous perturbation stochastic approximation by neighbor confidence (SPSA‐NC). Experimental results show that SPSA‐NC offers significantly better localization accuracy than its state‐of‐the‐art competitors, namely, simulated annealing and the ordinary SPSA. The SPSA‐NC program is available for downloading at http://www.dnagroup.org/SPSANC . Copyright © 2015 John Wiley & Sons, Ltd.

The recently emerging advanced metering infrastructure AMI is envisioned to be one of the most prominent features of smart grid. Security, especially authentication, is crucial for the success of large-scale AMI deployment. Unfortunately, AMI's natural requirements-efficiency, scalability, fault diagnoses, and reliability-cannot be fully satisfied by existing authentication schemes. To validate the delay-tolerant AMI metering data, we present a new authentication architecture that boosts the utilization of a set of efficient authentication schemes. Meanwhile, in case of authentication failures, fault diagnoses are highly demanded by AMI. Thus, we develop a few corresponding fault diagnosis algorithms that precisely pinpoint the authentication failure with lightweight computational cost. We implement our system on emulated smart meters and commodity servers. Experimental results on real-world scenarios demonstrate the feasibility of our system; it merely incurs substantially lighter overheads than those by the existing schemes, while it can effectively address the formidable authentication challenges in AMI. Copyright © 2014 John Wiley & Sons, Ltd.

To address recently emerging concerns on privacy violations, this paper investigates possible sensitive information leakages and analyzes potential privacy threats in the automatic appliance control (AAC) application, which is one of the handiest applications in smart grids and one of the earliest examples in Internet of Things (IoT). Without an effective and consistent privacy preservation mechanism, the adversary can capture, model, and divulge customers' behavior, activities, and personal information at almost every level of society. Based on a set of existing cryptographic primitives, we propose an attribute-based encryption (ABE) key management variant and we also design and implement a fine-grained protocol named privacy preservation protocol (P3). We further present a practical automatic appliance control (AAC) system based on that protocol, and shows that it can fulfill the smart grid's requirements in privacy preservation. Experimental results demonstrate that our protocol merely incurs a substantially light overhead on the AAC application, yet is able to address and solve the formidable privacy challenges both customers and utility companies are facing.

Localization precision remains active and open challenge in the area of wireless networks. For static network we develop model free approach of localization technique that by-passes the tedious modeling of diverse aspects to the contributing factor of localization errors, namely simultaneous perturbation stochastic approximation (SPSA) localization technique. The improved version of SPSA, simultaneous perturbation stochastic approximation by neighbor confidence (SPSA-NC) addresses error propagation of iterative localization controlled by incorporating a neighbor confidence matrix. The centralized SPSA and SPSA-NC does not scale well for the mobile environment due to the messaging requirements of repeated updates. We take distributed approaches to implement the aforementioned localization techniques for mobile devices by distributed simultaneous perturbation stochastic approximation (DSPSA) and distributed simultaneous perturbation stochastic approximation by neighbor confidence (DSPSA-NC) respectively, compare the results with the centroid (C) and weighted centroid (WC) localization techniques and show superiority of our methods.

Subspace clusters represent useful information in high-dimensional data. However, mining significant subspace clusters in continuous-valued 3D data such as stock-financial ratio-year data, or gene-sample-time data, is difficult. Firstly, typical metrics either find subspaces with very few objects, or they find too many insignificant subspaces - those which exist by chance. Besides, typical 3D subspace clustering approaches abound with parameters, which are usually set under biased assumptions, making the mining process a `guessing game'. We address these concerns by proposing an information theoretic measure, which allows us to identify 3D subspace clusters that stand out from the data. We also develop a highly effective, efficient and parameter-robust algorithm, which is a hybrid of information theoretical and statistical techniques, to mine these clusters. From extensive experimentations, we show that our approach can discover significant 3D subspace clusters embedded in 110 synthetic datasets of varying conditions. We also perform a case study on real-world stock datasets, which shows that our clusters can generate higher profits compared to those mined by other approaches.

Matching two 3-D trajectories is an important task in a number of applications. The trajectory matching problem can be solved by aligning the two trajectories and taking the alignment score as their similarity measurement. In this paper, we propose a new method called "TrajAlign" (Trajectory Alignment). It aligns two trajectories by means of aligning their representative distance matrices. Experimental results show that our method is significantly more precise than the existing state-of-the-art methods. While the existing methods can provide correct answers in only up to 67% of the test cases, TrajAlign can offer correct results in 79% (i.e. 12% more) of the test cases, TrajAlign is also computationally inexpensive, and can be used practically for applications that demand efficiency.

MOTIVATION: An important class of protein interactions involves the binding of a protein's domain to a short linear motif (SLiM) on its interacting partner. Extracting such motifs, either experimentally or computationally, is challenging because of their weak binding and high degree of degeneracy. Recent rapid increase of available protein structures provides an excellent opportunity to study SLiMs directly from their 3D structures. RESULTS: Using domain interface extraction (Diet), we characterized 452 distinct SLiMs from the Protein Data Bank (PDB), of which 155 are validated in varying degrees-40 have literature validation, 54 are supported by at least one domain-peptide structural instance, and another 61 have overrepresentation in high-throughput PPI data. We further observed that the lacklustre coverage of existing computational SLiM detection methods could be due to the common assumption that most SLiMs occur outside globular domain regions. 198 of 452 SLiM that we reported are actually found on domain-domain interface; some of them are implicated in autoimmune and neurodegenerative diseases. We suggest that these SLiMs would be useful for designing inhibitors against the pathogenic protein complexes underlying these diseases. Our findings show that 3D structure-based SLiM detection algorithms can provide a more complete coverage of SLiM-mediated protein interactions than current sequence-based approaches.

In this paper, we present a new scheme named "CP-Mine" for automatic three-dimensional (3D) protein structure classification using structural fingerprints. We represent a 3D protein structure as a CPset, which is a set of inter-SSE contact patterns (CPs) existing in the protein. Suppose we have a database of protein structures whose class labels are already known, and suppose there are distinct protein structure classes in the database. For each class, we generate its fingerprint by mining the frequent CPsets from all the member protein structures belonging to this class. When we want to predict the class label of an unknown protein, we also generate the CPset of this protein, and find the intersection between this CPset and the fingerprint of each protein structure class one by one. Then, the labels of the classes with the highest degree of intersection are returned as the answer. The proposed method is a pure classification scheme in that any kind of structural comparison, alignment or searching is not needed to be performed. The preliminary experimental results shows that our method can classify the protein structures accurately and efficiently.

MOTIVATION: As the sizes of three-dimensional (3D) protein structure databases are growing rapidly nowadays, exhaustive database searching, in which a 3D query structure is compared to each and every structure in the database, becomes inefficient. We propose a rapid 3D protein structure retrieval system named 'ProtDex2', in which we adopt the techniques used in information retrieval systems in order to perform rapid database searching without having access to every 3D structure in the database. The retrieval process is based on the inverted-file index constructed on the feature vectors of the relationships between the secondary structure elements (SSEs) of all the 3D protein structures in the database. ProtDex2 is a significant improvement, both in terms of speed and accuracy, upon its predecessor system, ProtDex. RESULTS: The experimental results show that ProtDex2 is very much faster than two well-known protein structure comparison methods, DALI and CE, yet not sacrificing on the accuracy of the comparison. When comparing with a similar SSE-based method, namely TopScan, ProtDex2 is much faster with comparable degree of accuracy. AVAILABILITY: The software is available at: http://xena1.ddns.comp.nus.edu.sg/~genesis/PD2.htm

In this paper, we present a novel indexing method called ProtDex to facilitate fast searching in 3-dimensional protein structure database. In ProtDex, we first build an index on the representative properties of all proteins in the database. When evaluating a query, with the help of the index, we filter out a small candidate list of proteins. Then, we can either directly report them, with their respective rankings, to the user, or do the expensive actual alignments on them upon user's request. Preliminary experimental results show that our solution is up to 16 times faster than the popular DALI method for database searching task (without actual alignments), while its overall accuracy is only slightly inferior to that of DALI. The software is available upon request by sending emails to the authors.